UNCLASSIFIED

AD NUMBER

AD479454

NEW LIMITATION CHANGE

TOApproved for public release, distributionunlimited

FROMDistribution authorized to U.S. Gov't.agencies and their contractors;Administrative/Operational Use; MAR 1966.Other requests shall be referred to U.S.Army Medical Unit, Fort Detrick, MD 21701.

AUTHORITY

USAMRIID ltr, 9 Jul 1971

THIS PAGE IS UNCLASSIFIED

U. S. ARMY MEDICAL UNIT

FORT DETRICK, FREDERICK, MARYLAND 21701

SPECIAL REPORT

TO THE

COMMISSION ON EPIDEMIOLOGICAL SURVEY

OF THE

ARMED FORCES EPIDEMIOLOGICAL BOARD

MARCH 1966

Project No. lB533001Dl64 03

COMPILED BY

HARRY G. DANGERFIELD, MAJOR9 MC

I'NMvI M ...S SANK, YWOU .. i i

ABSTRACT

A review of tke research program directed toward determining thebiological effect- of staphylococcal enterotoxin B is presented.

II

iv

FOREWORD

The papers presented here were given at the Executive Session of theCommission on Epidemiological Survey on 10 September 1965. Other uncl&sasi-fied reports were presented 9 September and were published separatelyFebruary 1966.

In conducting the research described the investigators &dheree. tn theprinciples of laboratory animal care as established by the National So:tetyfor Medical Research.

The technical assistance of Mrs. Phebe W. Summers in p-eparing theillustrations and editing the report is graLtfuliy aclrcwlcIdgcd.

DAN CROZIERColonel, MCCommanding

vm

TABLE OF CONTENTS

PAGE

Abstract i

Foreword ix

Pyrogenic Effects of Staphylococcal Enterotoxin B,

Frank A. Carozza, Jr., Captain, MC

Clearance and Localization Kinetics of Radioactive Labeled

Staphylococ4.al Enterotoxin B

Part I, Morton I. Rapoport, Captain, MC 14

Part II, Leland F. Hodoval, Captain, VC 29

Part III, Earl L. Morris, Captain, VC 35

Effects of Staphylococcal Enterotoxin B on the Coagulation Mechanism

and Leukocytic Response in Beagle Dogs --- A Preliminary Study,Charles F. Gilbert, Captain, MC 42

Serological Methods for Detection of Staphylococcal Enterotoxin B

Antibody,

Martha K. Ward, Captain, USPHS 49

Detection and Measurement of Immunological Responses toStaphylococcal Enterotoxin B.

Virginia G. McGann, Ph.D. 56

Distribution List 68

DD Form 1473 69

PYROGENIC EFFECTS OF STAPHYLOCOCCAL ENTEROTOXIN B

Frank A. Carozza, Jr., Captain, MC*

Recent evidence has indicated that staphylococcal enterotoxin B (SEB)exhibits a pyrogenic effect when administered parenterally to laboratoryanimals. Laboratory workers accidentally exposed to this material havealso become markedly febrile. The present investigations were carried cjtin an attempt to define more closely the febrile response to enterotoginin one species, the rabbit, and describe its mechanism of action.

Because fever and other physiologic changes, induced in experimentalanimals by SEB are similar to those induced by bacterial endotoxin,l / itwas necessary to demonstrate first that the pyrogenicity of enterotoxinwob not due to contaminating endotoxin. At the spring meeting of theCommission, evidence obtained in preliminary studies was presentcd indi-cating that pyrogenicity was an intrinsic property of enterotoxin. Thesedata are reviewed briefly. The possibility of endotoxin contaminationseemed unlikely since rigorous analysis of the SEB prepared and used atFort Detrick had revealed it to be an essentially pure substance. Never-theless bacterial endotoxins are ubiquitous and elusive substances;contamination of other materials with them is difficult to control. Infact, certain pharmacologic effects of endotoxin such as fever may becaused by trace amounts, hundredths of a microgram. To rule out the possibility that the pyrogenicity of SEB was due to contamination by traceamounts of endotoxin, a comparison was made of the febrile responses of2-kg albino rabbits to each of these substances.

In Figure 1 the solid line shows the composite febrile response of 6animals each given 10.0 pgikg of SEB intravenously (IV). The interruptedline represents the composite response of 5 normal rabbits to a single TVinjection of 1.0 pg/kg of Escherichia coli endotoxin. The area beneatheach curve is called the fever index and is a measure of pyrogenic activity.The fever indices for each group are almost identical over a 5-hr period:the shapcs of the curveb, however, are markedly different. The pyrogenicresponse to endotoxin was significantly higher at I hr than that producedby enterotoxin. At approximately 1 to 1i hr the response of endotoxin

recipients leveled off to form the initial part of the typical biphasicresponse. At 2 hr, temperatures of endotoxin-challenged animals began torise rapidly. A peak of 3.3 F above baseline temperature was reached by3 hr; over the following 2 hr the temperature rapidly feil to near normal.in contrast the temperature of enterotoxin recipients rose gradually and

by 5 hr reached a significantly higher mean temperature increment than theendotoxin group. In fact, peak increments did not occur until 6-7 1rfollowing challenge. The difference in the shapes of these two curves issignificant and suggested that the fever produced by enterotoxin was notdue to endotoxin contamination.

Medical Division.

m n • • • •

2

sEE{--a'IS.E.(6RBTS)0

ET -- - - ..E.(5RBT.S-z

SEB(IO.Oug/kg)

ET (I.Opg/kg e-U%z 4

clip

0 .I 50%

FI M- E (hr.

SINGLE IV INJECTION OF STAPHYLOCOCCAL

ENTFqOTOXIN B (SEB) OR ENDOTOXIN (ET).

3

if pyrogenic effects of enterotoxin had been due to endotoxin contami-natlon it should have been possible to render rabbits tolerant by repeatedinjcctions; in addition after repeated injectias of enterotoxin, rabbitsshould have been tolerant to a single injection of bacterial endotoxin.Figure 2 shows that this was not the case. A group of normal albino rabbitswas injected IV on day 0 with 10.0 pg/kg of enterotoxin. The mean 5-hr fevtrindex was 1062. This was not significantly different from the response of5 normal rabbits given 1 pg/kg of E. coli endotoxin, where the 5-hr tndcxwas 1106. When the enterotoxin group was rechallenged with the same doseA enterotoxin on days 3, 9 and 14, there was some suggestion of a decreasein mean fever index but this was not significant. On day 15, after 4 injec-tions of enterotoxin, these animals were given 1.0 pg/kg of endotoxin. Theresponse of these enterotoxin pretreated animals was approximately thatobserved in normal animals receiving the same endotoxin dose.

The converse of this expermcrnt was also performed. If the pyrogeniceftects of SEB were due to endotoxin contamination, animals renderedtolerant to endotoxin should have been refractory to the pyrogenic effectsof enterotoxin. Examination of Figure 3 reveals that tolerance to a largeamount of bacterial endotoxin did not confer tolerance to the pyrogeniceffects of enterotoxin. Rabbits were rendered tolerant by 5 IV injections0 f 10 pg/kg of E. coli endotoxin during a 9 day period. The mean feverindex on day 9 was 802. This reflected a high degree of tolerance since5 normal rabbits given half the dose of the same endotoxin had a mean feverindex of 1532. In spite of the high degree of endotoxin tolerance, rabbitschallenged with 10.0 lig/kg of SEB on day 12 reacted with as much fever asnormal animals receiving the same amount of enterotoxin. Thus endotoxin-.olerant a.iimals were not refractory to the pyrogenic effects of staphylo-coccal enterotoxin. The results of these studies indicated that pyrogenicitywas an intrinsic property of SEB and was not due to endotoxin contamination. I

Since these data were obtained, a further study has confirmed that pyro-genicity of SEB is not due to trace amounts of endotoxin. Although 4 injec.lions of enterotoxin over a 2-week period had failed to induce significantdimintution of the febrile response, rabbits were rendered pyrogenicallyrefractory to btaphylococcal enterotoxin by single daily injections (Figure 4).hIe bar on the left shows the mean pyrogenic response of 5 normal animals to10 ug/kg of SEB; the mean fever index is 988 and not significantly different!-.(m the response of normal rabbits to 1.0 pg/kg endotoxin shown at theextreme right. Enterotoxin was then administered by single daily iV Injec-tions. On day 4 the response to the 5th injection of 10 g/kg of enterotoxin,.s -educed significantly to 187. On the following day these animals, pyro--genically refractory to enterotoxin, were challenged with 1.0 Pg/kg endotoxinand responded as normals receiving the same dose. Therefore it is apparentrha- pyrogenic tolerance to endotoxin could not be induced with repeatedchalterges of enterotoxin even though after 4 daily enterotoxin injections.rl-nals became markedly refractory to its pyrogenic effects.

NORMALSEe 10.0mjg/kg (IV) E T (IV)

I I 11.0 mg/kg12 - _ _ _ _ _ _ __ _ _ _ _ _ _ E T (IV)

1.0 ug/kg

x 10

S8 1

NO. 66+ABIT

FIUEZ ERL EMS FRSIS R-RAEWIT 4TPYOOCLETRTXNB(E)

TO2~~ NOOI E)

I0

NORMAL RAB BITS

ET

16~~ E s eOv) -'tseB

ui 14 - 1 ong kg I. 'g\"g 1o g/kqa- 12 T E 1" us/kg (IV) ....,,, ..... O /k

tz -?- -f -

DAY 0 2 5 6 9 12NO. 6 6 6 6 6 6 5 5 6RABBITS

FIGURE 3. FEBRILE RESPONSE£ OF RABBITS, MADE TOLER-ANT TO E.. QWL ENDOTOXIN (ETr). TO STrAPHY-LOcoccATLENTEROTOXIN B (SEB).

6

. 4 NORMAL0 RABBITS

S . g/kg (IV) ET (IV) ET (IV)

+1 1.0 T 1"0 uIg/kg

x12xO I

z_8

w>6w

z2

20DAY 0 I 2 3 4 5

FIGURE 4. FEBRILE RESPONSE OF 5 RABBITS TO L 11ENDOTOXIN (ET) AFTER 5 DAILY INJECTIONSOF STAPHYLOCOCCAL ENTEROTOXIN B (SEB).

7

Having established that enterotoxin was intrinsically pyrogenic, studieswere performed to determine the mechanism of this effect. Because of thelatent period of k - 1 hr from the time of injection until an appreciablerise in the temperature of the rabbit, it seemed likely that an intermediatepyrogenic substance might be activated in vivo by the toxin. Furthermore,since enterotoxin affects leukocytes in vivo similarly to endotoxin,-' itseemed likely that this intermediate substance might be an endogenous pyrogensimilar to that which mediates fever after endotoxin administration. Figure 5shows that after enterotoxin administration an endogenous pyrogen is released.Six normal rabbits were given 10 pg/kg of SEB and the resultant compositetemperature curve is shown at the left. Four hr postinjection animals wereexsanguinated; the plasma was separated and frozen at -20 C. Because peaktemperatures occurred 6-7 hr after enterotoxin administration, 4 hr was feltto be a time when endogenous pyrogen would be maximal. Two days later thisplasma was assayed for endogenous pyrogen in rabbits rendered refractory toSEB by 8 daily injections. Recipient animals were rendered refractory inorder to negate the possibility that any pyrogenic substance in the donorplasma was merely enterotoxin. The broken curve on the right shows thecomposite response of the 6 recipients to 10 g/kg SIB on the day prior tothe endogenous pyrogen assay and reflects a high degree of refractoriness.When these animals received 15 ml/kg of donor plasma IV, a sharp incrementin temperature occurred as shown by the solid line on the right. This feveroccurred promptly, peaked to about 1.5 F above baseline and disappeared by3 hr. Such a response is typical 7f that caused by the endogenous pyrogen Ireleased by bacterial endotoxins.1 / Enterotoxin-refractory rabbits whoreceived normal donor plasma showed no response.

A similar study further confirmed that the release of endogenous pyrogenby enterotoxin is not due to endotoxin contamination. Donor animals werepretreated with daily injections of E. coli endotoxin for I week prior toenterotoxin administration as shown at the left in Figure 6. The monophasic

temperature curve represents the endotoxin-tolerant state. These animalspresumably had no circulating endogenous pyrogen 4 hr after endotoxininjection. The remainder of this study was carried out as described previously.Four hours after enterotoxin administration animals were bled; plasma wasassaved in enterotoxin refractory recipients for endogenous pyrogen. A prompt.onphasic fever similar to that seen before ensued. Thus staphylococcalenterotoxin does induce an endogenous pyrogen in recipient animals, whichpiobably mediates the febrile response.

It was stated earlier that after 4 single daily IV injections of SEBtabbits become refractory to its pyrogenic effects. The remainder of thisreport is concerned with a further examinaticn of this state and a prelimi-nary investigation of mechanisms responsible for it.

Figure 7 shows the response of 4 normal animals to 10 pg/kg enterotoxin,the curve labeled day 0. The mean 6-hr fever index of 1,000 is shown on theright. On the next day when rechallenged with the same dose there was aslight, but not significant, decrease in the mean fever index. Single daily

i

NORMAL DONORS (6)1 SEB REFRACTORY RECIPIENTS (6)

- SEB IO.Opq/kg IV.---*14-

.. fSEB 10.0 faq/kq IVBLE-

0 PLASMAz (E.P.)

W 215 m1/kqIV -

CLaI -

0 I 2 3 40 I 2 3TIME (HOURS)

FIGURE 5~TRANSFER OF ENDOGENOUS PYROGEN INDUCEDBY SES IN NORMAL ANIMALS.

ENDOTOXIN TOLERANT SEB REFRACTORYDONORS (7) RECIPIENTS (4)

-4

:. COLI ENDOTOXIN 1.0 pq/kg IV f-SE8 10.0 yqlkg IVO-

5- SEB 10.0 pq/kg IVBLEED1

(.1 PLASMA2 ooi (E.P.)

IS mi/kg IV.-

I-

a.

0 1 2 354 0 1 2 3 4 0 I 2 3TIME (HOURS)

FIGURE 6. TRANSFER OF ENDOGENOUS PYROeEN INDUCEDBY SEB IN ENDOTOXIN -TOLERANT ANIMALS.

10

SEB 10.0 pq/kg IV+ 4 4 1 4 12..

Z I

w 4- 1.6SEB 10.0 paq/kq IV x

z zTM (DAYO) Y

m- w

AND 2O'DAWING 0 (DAY 7)ST.

(DAY 7)

Iyw ~ *(DAY 4)

w, * Lv

012 3 456 0 I 2 4 5 6 71TIME (HOURS) DAY

FIGURE 7 PYROGENIC RESPONSES OF 4 RABBITS TOREPEATED SEB ADMINISTRATION FOR 5 DAYSAND FOLLOWING 2 DAYS REST.

Injections were continued and by day 4, the 5th challenge, animals werepyrogenically refractory. Animals were not challenged with enterotoxinon the subsequent 2 days. However, after this period of rest rabbitschallenged with the same dose eAhibited an almost normal febrile responseto enterotoxin with a fever index approaching that observed after the 1stinjection. Thus the pyrogenic refractory state induced by 4 injectionsof enterotoxin was transient and almost completely abolished by 2 days rest.

Studies were then initiated to determine the mechanism of the entero-toxin pyrogenic refractory state. Its transient nature suggested that theanimals might be nonspecifically refractory due to loss of sensitivity ofthe thermoregulatory centers to pyrogenic stimuli or inability to furthermobilize endogenous pyrogen. Because refractory animals responded normallyto passive transfer of endogenous pyrogen and to bacterial endotoxin, how-ever, this possibility seemed unlikely.

Another possibility was that refractory rabbits might be protected fromthe pyrogenic effect of enterotoxin by a circulating humoral factor or anti-body. No antibody, however, could be detected in control or refractorystate sera employing the agar diffusion test or tanned sheep cell hemaggluti-nation test.

Nevertheless, the possibility of a circulating protective factor, notmeasured by the serologic tests employed, remained. Therefore an attemptwas made to demonstrate such a factor by means of passive transfer of therefractory state. Figure 8 shows that this could not be done. Eightrabbits were rendered pyrogenically refractory to SEB by 4 daily injections.The pyrogenic reaction to the 5th challenge is shown on the left. On thefollowing day animals were bled; plasma was separated and frozen. Normalrabbit plasma was also collected and similarly treated. Two weeks later 6 jnormal rabbits were given IV 15 ml/kg of refractory state plasma and 6,15 ml/kg normal plasma. One hour postinjection all animals were challengedwith 10 Pg/kg IV of SEB. Results are shown to the right. The mean 6-hrfever index did not differ significantly between the 2 groups receivingdonor plasma. Moreover, the mean fever index for each group was the sameas normal animals receiving the same dose. Thus, no circulating protectivefaztor could be demonstrated in animals pyrogenically refractory to SEB.

Other mechanisms, which may mediate this state such as hyperactivity ofthe reticuloendothelial system and specific desensitization to enterotoxinare presently under investigation.

SUMMARY

Pyrogenicity is an intrinsic property of staphylococcal enterotoxin B.

Upon intravenous administration in the rabbit, enterotoxin releases anendogenous pyrogen.

12

COMPOSITE FEBRILE SEB 10.0 jiq/kg IVRESPONSE OF 6 PYROGENICREFRACTORY DONORS TO

51SEB. NORMAL PRETREATMENT cj.

NORMAL REFRACTORt 13 2*DONOR DONOR x

4- PLASMA PLASMA -12 xT w

SE 10.0 jq/ko IVa

wo>z w'A.

z

w1-(DAY 4) A-

0 123 4 66TIME (HOURS) NO. RABBITS

FIGURE 8. PYROGENIC EFFECT OF SEB ON RABBITS PRE-TREATED WITH PLASMA FROM PYROGENICREFRACTORY DONORS.

13

Rabbits become pyrogenically refractory to enterotoxin after 4 singledaily intravenous injections. This state is transitory and disappearswithin 2 days if animals are not repeatedly challenged. The refractorystate is probably specific for enterotoxin since refractory animals reactnormally to endotoxin and endogenous pyrogen. The refractory state isnot mediated by a demonstrable circulating humoral protective factor.

LITERATURE CITED

1. Sugiyama, H., E. M. McKissic, Jr., and T. Hayama. 1964. Hyperfibrino-genemia and thrombocytopenia after staphylococcal enterotoxin. Proc.Soc. Exp. Biol. Med. 117:726.-729.

2. Atkins, E. 1960, Pathogenesis of fever. Physiol. Rev. 40:580-646.

3. Petersdorf, R. G., and I. L. Bennett, Jr. 1959. The experimentalapproach to the mechanism of fever. Arch. Int. Had. 103:991-1001.

A

I

II

I

Ii

14

CLEARANCE AND LOCALIZATION KINETICS OF RADIOACTIVILABELED STAPHYLOCOCCAL ENTEROTOXIN B

PART I

Morton I. Rapoport, Captain, MC*

Detailed physiological data obtained as a result of investigations intothe mechanism of action of staphylococcal enterotoxin B (SEB) have beenpresented in a previous session of this Commission. Despite the chemicaldissimilarity of SEB and bacterial endotoxin, they do share some importantphysiological responses. In addition to fever, both are capable of producinghypotension, avotemia and leukopenia and alterations of glucose metabolismand catecholamine response. In contrast to these similarities, the authorpresented data at the last session of the Commission that suggested thatSHB was reversibly bound to tissues and that this phenomenon was importantin defining its mechanism of action.

This binding concept became apparent with the use of the highly specificequine origin antibody, staphylococcal enterotoxin B antitoxin Lot 1, pre-pared by Lt Colonel GroganY We have continued to use this antiserum as ameans of learning more regarding the mechanisms by which SEB is lethal. Asa result, we have gathered significant additional information by a varietyof approaches all of which support the concept of reversible tissue bindingas a factor in toxemia and provide a valid approach toward therapy. Datapresented in April 1965 are reviewed. Results of recent studies on theinteraction of SEE with the reticuloendothelial system (RES) and specificbinding sites in tissues is presented. This work permits the bindingconcept to be considered in a much more detailed fashion.

The interaction of toxin and tissue sites first became apparent as aresults of clearance studies. SEE can be tagged easily with radioiodine(1131); subsequent work indicated this rag was relatively stable. Theisotope appeared to have no demonstrable physiological or immunologicaleffect on the toxin.

Figure 1 shows the clearance from plasma of SgB-1 131 in a group ofmonkeys given 1,000 pg/kg (approximately 30 LDs0) of body weight. Thesolid line indicates total plasma radioactivity and the broken line,toxin activity. The latter was determined by trichloroacetic acid (TCA)precipitation of protein bound radioactivity. Throughout this discussionprotein bound radioactivity will be equated with toxin radioactivity sincespontaneous adherence of free 1131 to serum proteins is comparativelyminimal. Clearance is plotted as per cent of initial counts per minute(CP1).

* Physical Sciences Division.# Assistant to the Commanding Officer.

15

Disappearance of toxin from plasma was exceedingly rapid, with a half-disappearance time (t ) of 7 min. During the inital 60 min, total plasmaradioactivity appeared to reflect accurately toxin activity. Thereafter,delodination, indicated by the separation of the curves, progressed at arate sufficiently rapid so as to render a measurement of total plasmaradioactivity misleading. The slight rise in total plasma radioactivityafter 3 hr represents the recirculation of free 1131.

In contrast are the results shown in Figure 2 from a group of monkeysgiven type-specific antitoxin prophylactically 30 min prior to challenge;a marked delay in clearance of SEB-I11 was evident, the tj being > 5 hrrather than the previously observed 7 min. The rate of de odination oftoxin was also inhibited as demonstrated by the minimal separation of the' curves. The toxin was held in the circulation and was metabolized,presumably very gradually, or altered in a manner which prevented it fromcausing death.

In view of earlier work indicating a reversal of toxemia with delayedadministration of antitoxin, exeriments were designed to test the effect,if any, on clearance of SEB-I 3 when antitoxin was administered thera- Ipeutically.

Figure 3 shows the 13ffect of different quantities of antitoxin adminis-tered 2 hr after SEB-I challenge. The open circles indicate a group of4 animals given antitoxin, 6 ml/kg of body weight, and the closed circles4 animals given 3 ml/kg. Toxin removal from the circulation was not onlyhalted but there was a rather pronounced return of radioactivity to theplasma. While the differences were not statistically significant due tothe relatively small number of animals studied it appeared that the higherdose of antitoxin returned a larger amount of radioactivity to the circu-lation; returned radicactivity was almost exclusively toxin radioactivityas indicated by the broken lines. In addition, the return of toxin tothe circulation was even more pronounced when animals were given antitoxin30 min after challenge.

Jt was clear that despite rapid removal of SEB from the circulationits did not imply detoxification since animals given type-specific anti-

body prophylactically, were protected but, paradoxically, cleared SEB ata much slower rate. One might conclude that animals not receiving antitrxin cleared the material rapidly with the result that the toxin estab-fished itself upon or within vital sites, whereas, toxin did not reachtilose vital sites in animals receiving prophylactic antitoxin. Furthermore,delayed administration of antibody reversed the clearance pattern andreturned a portion of the toxin to the circulation. This finding suggestedthat SEB was bound or attached to certain sites from which it could beremoved and concomitantly, toxicity reversed.

I16

Io

IG

40

-4 TOVaLAPIASLA RAStOACTIVITT

to ** PLASMA TC 1 c MIPIATt

1010

a so

go

40

30 60 110 160 240 0 M 4 0

TIMs (MI*IuT 3)

FIGURE I. DISAPPEARANCE OF SEB-1131 IN 6 MONKEYS.

100

ANIOXI RO OCALNEMe:

_j 40' w-...TOTAL PLASMA RADIOACTIVITY

1--, PLASMA TCA PRECIP[TATE

20

3060 20 60 140 300 4 420TIN (MINUTES)

FIGURE 2. DISAPPEARANCE OF SEB- II1 IN 4 MONKEYS GIVEN

ANTITOXIN PRIOR TO CHALLENGE.

.oo

00

" , TOTAL TCA OPT i€5I41 PJ llASMA I'

M

1140 t 1410 140 I00 140 4110TIM( 435I5UT00)

FIGURE . REAPPEARANCE OF SEEB-I13 I AFTER ADMINISTRATIONOF ANTITOXIN AT TWO DOSAGE SCHEDULES.

i

17

The concept is presented in equation form in Figure 4 The very rapidremoval of SEB from the circulation implies that the number of potentialtissue binding sites far exceeds the number of toxin molecules administered.

Figure 5 shows that the addition of type-spe!ific antibody reverses theequilibrium; so that less SZB is bound to tissue sites. More simply stated,the binding affinity of SEB for antibody seemed to exceed the binding affinityof SEB for tissue sites. This concept seemed most attractive and additionalstudies designed to test this hypothesis are reported.

In an attempt to gain knowledge regarding interaction of SEB and aspecific tissue site, and in addition, to measure a response to physiologicalstress, we selected adrenal cortical function as a study model. Bacterialendotoxin has been investigated extensively in regard to adrenal physiologyand found to induce an adrenal glucocorticoid response. This observationhas been a sou-ce of valuable information both from a mechanistic as wellas therapeutic aspect.

Values of plasma Porter Silber chromogens in monkeys under severalexperimental conditions are shown in Figure 6. Each curve represents agroup of monkeys bled serially during a 24-hr period (number of animalsstudied are shown in parentheses). To avoid iatrogenic hypovolemia duringthe study each bleeding was immediately followed by replacement of an equalvolume of banked whole monkey blood. Challenge materials were administeredimmediately after the 0 time bleeding.

The broken line represents values from a group of control monkeys givensaline intravenously (IV). Plasma corticoid concentrations showed only theslight fluctuation to be expected in a normal diurnal pattern: this impliedthat the handling and bleeding procedures were not in themselves sufficientlystressful to evoke an adrenal response. The dotted line indicates a groupof monkeys each of which received 4 units/kg of adrenocorticotropin (ACTH)in single IV injections. The maximum stimulation occurred at 4 hr with themean value being 81 pg/100 ml; thereafter, plasma hydroxycorticoids returnedto control concentrations. The 3rd group of animals received IV 100 pg/kgof SE3; the values are indicated by the solid line. Following challengethere was a progressive rise reaching a peak of 127 pg/lO0 ml at 4 hr. Itis apparent that SEB in the dose administered was significantly more potentir st'm-ilating adrenal corticoid secretion than an acute single injection,f 4 units of ACTH, a dose in the physiologic range.

Additional experimental variables were studied in the same fashion,as shown in Figure 7. The solid line indicates 17-OHCS values in bloodfrom a group of animals given SEB. The broken line represents 6 monkeys

given antitoxin prophylactically 30 min prior to challenge with SEB. Thereis only a modest rise in cortisol values in this group. In contrast,another group of 6 monkeys were given antitoxin therapeutically 30 minafter SEB-. the rise in 17-OHCS was intermediate between groups treatedprophylactically and those remaining untreated.

SEB + TiSi SEB-Ti Si? SPECIFIC OR

NON SPECIFICBINDING?

FIGURE 4. EQUILIBRIUM OF SEB WITH TISSUE BINDINGSITES.

KSEB+4 TISi SEB-TiSI

SEB-AbFIGURE 5. EFFECT OF TYPE SPECIFIC ANTIBODY ON

EQUILIBRIUM OF SIB WITH TISSUE BINDINGSITES.

19

...... .... .... " ;; .......

I-.

FIUR . MOKE PLSM -01I

.... .... . .... .

FIGURE 6, MONKEY PLASMA I7-ONCS.

FIGURE 7. MONKEY PLASMA I7-OHCS. iII

i

20

The experimental design did not permit recognition of any exact site,such as the cerebral cortex, hypothalamus or pituitary, where the toxinmight initiate adrenal stimulation. Previous work had indicated a pro-nounced catecholamine release following SEB challenge; epinephrine is awell known stimulator of ACTH. However, SEB is a considerably largerprotein than the various octopeptides, known to release ACTH. The impli-cation from this study is that antitoxin lessens adherence of toxin, orremoves it from, the sites that influence adrenal secretion.

It became important to define the role of the RES in host defenseagainst SEB intoxication with respect to tissue binding. It is doubtfulthat a substance initially phagocytized by the RES would remain immuno-logically active and capable of being released by addition of antibody.

Figure 8 taken from an article by Iio and Wagner shows clearance ratesof radioiodinated aggregated albumin from plasma of dogs. The solid linerepresents total radioactivity and the broken line, protein bound activity.It may be seen that clearance is rapid; within 12 min a reappearance ofradioactivity occurs. This secondary rise is a function of the release ordisassociation of 1131 from the aggregated albumin. Protein bound 1131continues to leave the plasma in a progressive fashion.

Results of similar work conducted in the Medical Unit in a group ofmonkeys are presented in Figure 9. Radioiodinated aggregated albumin wassupplied by Squibb Laboratories. It is apparent that the clearance ofalbumin given in a dose of 3 mg/kg is relatively rapid as is the rate ofdeiodination. In contrast, a group of monkeys given Thorotrast, 3 ml/kg,3 hr prior to administration of aggregated albumin, is shown on the right.The clearance rate is markedly inhibited and the rate of deiodination isdelayed. This procedure then, defined a change after Thorotrast in uptakeand deiodination of a substance known to be phagocytized by the RES.

A similar study performed in monkeys receiving SEB-I1 31 is shown inFigure 10. The disappearance of SEB in the control situation is shown onthe left. The clearance of SEB-I 1 31 in a group of monkeys given Thoro-trast 3 hr prior to challenge is shown on the right. It is apparent thatthe rate of clearance and the rate of delodination are uneffected by thismanipulation. The implication is that the RES must play only a minor rolein the clearance of this material. This conclusion is also supported bythe observation that type-specific antibody actually delays removal of SEBrather than facilitating it, as is the case with bacterial endotoxin.

The fact that the RES does not appear to be a major factor in SEBintoxication is in keeping with our observation that delayed antitoxintherapy seemed to remove toxin from unidentified sites and return it tothe circulation. Immunoactive or unaltered toxin would not be expectedto return from intracellular sites within the RES.

21

Colloidal Albumin Disoppeovonce Rajte Constant

29.6*41.2%mmiColloidalAbmnDerdto Rote Constow,

0 4erdtnII0%*i

06

T Io0.1- T T T T

0.2 24A.4 G

004

FIGURE S. BLOOD CLEARANCE Of AGGREGATED ALBUMININ NORMAL DOGS.- TOTAL PLASMA RADIOACTIVITY.- -PLASMA RADIOACTIVITY, AFTER REMOVALOP FREE IODIDE.(Aft.r tio st *t. J. Cliii. Invest, 42,417.1963, Reprinted withpermission of the Jovi not of Clinica Investtion)fl

4

22

% INITAL CPM

00 0 0 00 0

0 -

r -4

~~30

I"0

-0

4L4

41 00

I-

w p1.- 00

g . p cuJ

U ~10

A U)

00

0

J..

Aa.

OD (a

I

24The observation that the R3S is not a major removal organ forSEB

leaves the obvious question: what are the sites of importance in theremoval of SEB? Dr. Morris' work (presented later in this report) indi-cates that a large amount of SIB radioactivity localized in the kidneys.This finding was in agreement with earlier work done by Lt Colonel Dirks(Pathology Division, Medical Unit) using an entirely different experi-mental approach, fluorescent antibody staining of kidney sections. Figure11 is a black and white reproduction of one of his slides. A pronouncedlocalization of SIB within the proximal tubular epithelial cells of thekidney was noted. The light area fluoresced apple green and denotedpresence of SIB.

It was observed further that antibody prophylaxis drastically alteredlocalization in the kidney so that only small amounts of activity accumu-lated. These findings, in conjunction with previous work indicating amarked reduction in the rate of deiodination as a result of antitoxin pro-phylaxis, prompted rurther investigation of the renal aspect of the problem.

The clearance of SIB-1 131 in a group of monkeys which had bilateralligations of the renal artery and vein 1 hr prior to challenge, is shownin Figure 12. There was an obvious delay in disappearance of toxin, thetk approximated 30 min rather than the expected 7 min, DeLodination wasmarkedly inhibited; there was no evidence of the secondary rise ?f radio-activity attributed to the appearance of free or disassociated 1 31. Thestudy indicated that the kidneys were not only a major site of locali-zation but also a primary site of deiodination and perhaps other metabolicalterations. Furthermore, the prophylactic administration of specificantitoxin prevented localization of toxin in the kidney and perhaps, as aresult, inhibited its deiodination.

It was appArent that despite numerous metabolic similarities withbacterial endotoxin, SIB exhibited very obvious differences in localizationsites and clearance kinetics: the RES did not appear to play a major rolein the removal of SIB from the circulation and work indicated that thekidneys were clearly important organs of metabolism. Whether the oliguriaand azotemia associated with SIB intoxication were due to direct effect ofthe toxin or merely secondary to shock was less evident.

The reversible binding concept remains the most attractive explanationfor the sequential changes observed.

Figure 13 illustrates in a slightly more detailed manner than thatshown in Figure 5, our present concept of the distribution kinetics of SEB.The primary organs of localization are the liver and kidneys. The latterare the major sites of deiodination of labeled SIB and perhaps the majorsite of metabolic alteration. It was further shown by Dirks by means ofthe fluorescent antibody technique that SIB accumulated in the brain aftera delay of 2-4 hr. Adherence of SIB to white blood cells in vitro has beenobserved but the significance has not been established. The RES isapparently a minor site of localization and metabolism.

25I

FIGURE 11. LOCALIZATION OF SEE IN RENAL TUBULAREPITHIELIAL CELLS DEMONSTRATED BYFLUORESCENT ANTIBODY TECHNIQUE.

j

I

26

% INITIAL CPM

*h wo-0 0 0 0 00

A"

FF

0 /0

lip

,4 r"

z _0

0

K /

-1

27

LIVER MUSCLE

//SEB + Tissue Sites ----- KIDNEY

RES WOC LUNG

LIVER MUSCLE/ CU

Ab + SEB + Tissue Sites =KIDNEY

RES WUC LUNG

FIGURE 13 & 14. DEMONSTRATION OF EQUILIBRIUM REVERSALRESULTING FROM ANTIBODY ADMINISTRATION.

I

i

II

28

Figure 14 indicates the sequence of events occurring after antibodyadministration. The reversible binding or equilibrium relationship wasmost prominent in the kidneys from which a large amount of SIB apparentlywas returned to the circulation. This was expiained by the binding

affinity of SIB for antibody exceeding its affinity for the kidney. To

a lesser extent this sequence occurred also in liver and muscle. Curiouslyenough, SIB was increased in the lung following antibody administration.The explanation for this observation was not apparent although this most

probably represented accumulation of the toxin-antibody complex rather

than toxin alone.

SUMMARY

There is a suggestion that following SEB challenge, toxin rapidly

binds to certain vital tissues and in this manner manifesti its toxicity.The fact that this binding can be reversed to a dehree servee as a basis

for the effectiveness of delayed therapy.

Considerable insight has been gained into the mechanisms by which SIB

acts. Productive leads are being pursued with emphasis on the metabolic

and physiologic interactions of toxin and host.

I

I

29

CLEARANCE AND LOCALIZATION KINETICS OF RADIOACTLVELABID STAPHYIOCOCCAL RNTXROTOXIN B

PART II

Leland F. Hodoval, Captain, VC*

The theory of a reversible binding phenomnon between staphylococcalenterotoxin B (SEB), its specific antitoxin, and various undetermined tissuesites is expanded by description of a radiobioassay and its implications.

Evidence was presented by Captain Rapoport that antitoxin injected intothe circulation 30 min after the administration of radio-iodinated SEB(SEB-I131) caused a reappearance of labeled toxin in peripheral blood. Inpassively immunized monkeys the disappearance rate was prolonged from anormal of 7 min to approximately 5 hr.

Following these observations, it became evident that the binding betweenthe toxin, its specific antitoxin, and undetermined tissue sites resembledthe reversible binding phenomenon characteristic of several hormones andtheir specific carrier proteins.

It was theorized taat retention within the blood stream (half-dis-appearance time or t ) of a given dose of toxin would be increased if thequantity of circulating specific antitoxin were increased. Thus, it seemedreasonable th..t SEB protective antibody might be quantitated by its abilityto prolong the disappearance rate of the SZB-i131 in an experimental animal.This report describes the development of such a bioassay using albino rats,and demonstrates its potential usefulness in the study of serum from avariety of animal species including humans with hesagglutinating (HA)antibodies against SEB.

White male rats weighing 250-280 gm were used throughout. Since pre-liminary studies failed Lo reveal evidence of any influence of rat weight jwithin this range, a single uniform dose of 360 pg of SEB-11 31 was used.

Radtioactive toxin and specific antitoxin were identical to thoseprevio,,ely described. For purposes of this assay, the standard, undilutedhyperi.--une equine antiserum with an HA antibody titer of 1:1,600,000 wasassigned arbitrarily a value of 16.0 units/ml.

Figure I illustvates schematically the bioassay protocol. Followinganesthetization of a rat by an intraperitoneal injection of 0.6 ml of 5%sodium pentothal containing 120 units of heparin, exactly 0.5 ml of thestandard antiserum in appropriate dilution, or unknown serum to be assayed,was injected into the dorsal penfil vain. One minute after injection ofthe test sample, 360 Vg of SEB-I was injected into the same vein.* Physical Sciences Division.

30

Blood samples for determination of radioactivity were obtained from

the tail artery; the tail was cut with large scissors and clamped.Samples were taken at precisely 4, 6, 8, 10, and 14 min after toxininjection. Droplets of blood were collected on a small piece of parafilmand the tail stump was reclamped. The first sample was obtained at mid-tail and subsequent samples, immediately proximal to the clamp. Followingthe last collection the rat was sacrificed.

The whole blood was aspirated immediately from the parafilm with a0.l-ml micropipette and blown into the bottora of it plastic tube forcounting. Two ml of distilled water were flushed through the pipette andinto the counting tube to remove any residual blood. Radioactivity wasthen measured in a scintillation well detector. The counts obtained wereplotted against time on semilogarithmic paper and the t was determined.Figure 2 shows examples of such plots in individual animals. As can beseen, essentially linear disappearance curves were obtained.

Figure 2A shows an example of a control rat given 0.3 ml saline asa test sample; tk was 3.8 min. The average tk of 10 such control animalswas 3.7 t 0.2 min. Figures 2B, 2C, and 2D show that rats passivelyimmunized with increasing amounts of antitoxin demonstrated a progressiveprolongation of th. In Figure 2B, 2.0 units of antitoxin were administered;a tk of 10.8 min resulted. Figure 2C shows a th of 23.7 min with 4 unitsand Figure 20, 34.0 min with 8 units of antitoxin.

When the half-disappearance time in individual animals given anidentical amount of toxin was plotted against the logarithm of the unitsof antitoxin a sigmoid dose response curve was obtained (Figure 3). Con-trnl animals and 8 concentrations of antitoxin were assayed. It can beobserved that the range of values for an individual dose of antitoxinbecame widest with the highest concentrations. This could be attributedperhaps to the flattened disappearance curve in individualirats whichtended to decrease the accuracy of tk estimation; it could also be relatedto increased inherent variability in the region of flattening of the assaycurve itself.

it seemed appropriate to determine if further increases in antitoxinconcentration would influence the bioassay curve. In order to avoidincreasing the test sample volume, purified and concentrated hyperinmuneequ'ne SEB antitoxin (The National Drug Co., Lot 19575) was employed.With undiluted samples of this, the tj was increased to 45 min. Theresults of assays of serial dilutions of this product suggested that itwas '-4 times more potent than the antitoxin used for the standard. Fromthese observations it was concluded that the dose response curve wouldflatten progressively with the administration of antitoxin in doses above8.0 units/0.5 ml.

The portion of the dose response curve which showed the sharpestlinear increase fell between 3 and 5 units. Therefore, this portion ofthe curve was studied in greater detail (Figure 4). Thirty-five animals

31

ANESTHETIC3

SV SM LE NjECTIONS

SLOOD SAMPLI

-j0 4 6 a 0 64MvIUTEs

FIGURE I. SEB ANTITOXIN BIOASSAY PROTOCOL.

to 004 to to0

to oIo

La .

0 4 0 iW 64 to 6 4 6 11 l6 to

FIGURE 2. SEB-I131 DISAPPEARANCE IN RAT..o1

30

S C , 2o .C 4O sc

UNITS SEBA

FIGURE 3. SEBA ASSAY- ENTIRE CURVE. III

1

32

40 1

UNITS -si

FIGURE 4. SEBA ASSAY- LINEAR PORTION.

FIGURE 5. SEBA ASSAY - HUMAN SERA.

33

were employed using dilutions of the standard in the linear portion ofthe curve. The best-fit line, as shown, was obtained by the least squaresmethod and exhibited a correlation coefficient of 0.99. A linear relation-ship was confirmed by an analysis of variance and was significant at < 0.01%level. The lambda value (criterion of likelihood) for use of this portionof the curve for assay purpose was 0.132. It should be noted that theabsolute location of this curve varied slightly with different lots ofSEB-1 13 1 and accordingly, unknown sera should always be compared with anti-toxin standards. In addition, unknown sera with prolonged half-disappearancetimes would have to be diluted as necessary to permit quantitation within thelinear portion of the assay curve.

To test the assay, sera were obtained from normal and immunized horses,burros, and monkeys. The normal sera from these species all showed valuessimilar to those obtained with normal saline controls. Assays of sera fromthe immunized animals showed a prolongation of the disappearance timeenablivg an estimation of their potency to be made.

Sera from 25 normal human subjects were studied. Although none of thisgroup gave a history of known exposure to SIB, 19 had HA antibody titersranging from 1:1,000-1:32,000. The average tk of duplicate samples of thesesera is plotted against HA titer in Figure 5. None was found different fromthat of saline controls. An interesting observation on human gamma globulinobtained commercially was that it had an HA titer of 1:82,000 and caused aprolongat.jn of t equivalent to 2 units of antitoxin (note the asterisk onFigure 5).

By interpolation, one antitoxin unit was equated with a 1:100,000 HAtiter. As can be seen, the lower limit of sensitivity of the bioassaysystem appeared to be in keeping with an HA titer of about 1:50,000.

If our interpretation of the available data approaches a true inter-relationship, it implies that the rat bioassay is much less sensitive thanHA titers for detecting evidence of an immune response to SEB. But it doesimply that very high HA titers must exist in serum before the t of SEB islengthened.

The theory upon which our assay was based would imply that similarresults could be obtained in other species. To test this hypothesis,the tk of radioiodinated toxin was determined in guinea pigs and found tobe approximately 7 min; this was lengthened progressively in individualanimals by the prior administration of antitoxin in increasing concen-trations indicating that a similar pattern existed in other species.

The deomonstration that a reproducible dose-response relationship existedbetween antitoxin and the rate of toxin clearance gave support to our basictheory concerning such an interaction. The findings observed in rats andguinea pigs, together with those in motakeys, implied that such a relationship

34

was a general one and not species-specific. That such a doss-responserelationship could be predicted added value to our supposition thatbinding phenomena could explain the interaction between BD and itsspecific antitoxin.

The specificity of this assay appears to depend entirely upon thepresence of antitoxin against 39B in the material tested. The conceptthat the antitoxin-toxin combination is retained within the circulationand therefore is prevented from reaching sites within the tissues toexert toxiv action is based upon a single mechanism. This concept impliesthat protetive activity and half-life-lengthening activity are one andthe same property of the antitoxin. Whether the antitoxin molecules whichafford protection are identical to those which hold the toxin within theblood stream cannot be determined from the information at hand.

Less well defined is the relationship between HA activity and theability of the antitoxin to protect with or without lengthening of thetj of the toxin.

SUMARY

A rat radiobioassay for SEB antitoxin provided reproducible resultsof high precision. This systom has given support to our concept con-cerning the toxin-antitoxin interaction.

35

CLEARANCE AND LOCALIZATION KINETICS OF RADIOACTIVLABELED STAPHYLOCOCCAL ENTEROTOXIN B

PART III

Earl L. Morris, Captain, VC*

Fundamental to obtaining an understanding of the pathophysiology ofstaphylococcal enterotoxin B (SEB) intoxication is the location of thesite or sites of localization of the toxin. Early studies of toxindistribution were conducted in monkeys 2, 6, 12 and 24 hr postchallengewith radioiodine-labeled SEB (SEB-I1 31). However, it was subsequentlydetermined that over one-half of the SIB-I 131 was deiodinated within4-6 hr postchallenge.

In the present study, Macaca mulatta without known prior experiencewhile unf sodium pentothal anesthesia, were challenged with 1,000 .g/kgof SEB-I , a dose expected to produce lethality in 80-90% of monkeys.Ten minutes and h, 2 and 6 hr later they were sacrificed, autopsied, andorgans to be counted were placed in plastic cups; the carcass was retainedfor counting. Radioactivity was measured in a total-animal counter whichmaintained a constant counting geometry in order to quantitate and localizethe activity and account for 100% of the injected radioactivity. An attemptwas made to determine if there was a variation in tissue distribution dueto route of administration and to show the influence of antitoxin on thedistribution of SEB.

Table I lists the routes of challenge, number of animals employed, andthe intervals at which they were sacrificed. The antitoxin employed washyperimmune equine serum (SERA) and was administered intravenously (IV)J hr prior to challenge. Following antitoxin prophylaxis virtually allof the SEB remained in the blood stream 10 min postchallenge; therefore,this time period was eliminated in the animals so treated.

The ability to study localization ofSPxin following oral challengedepended upon the stability of the SEB-I complex following absorption

through the intestinal wall. An iyJial study was directed toward thisproblem. Two monkeys were given I -labeled toxin intragastrically and2 were given free 1131 by the same route. All were bled at frequentintervals over a 6 hr period. Each blood sample was centrifuged and 1 ml

of plasma and the precipitate obtained from treating 1 ml of plasma withtrichloroacetic acid were counted. The results are shown in Figure 1.

In the group given SEB-I131 the majority of radioactivity was notbound to protein. Essentially the same quantity of protein-bound radio-activity was found in both controls and challenged animals. This showed

* Physical Sciences Division.

36

TABLE I. DISTRIBUTION OF MONKEYS IN EXPERIMENTAL PROTOCOLS

CHALLENGE SEB-I 1 3 1 VARIABLES NO. OF ANIMALS BY TIME (min)

ROUTE 10 30 120 360

Inferior vena cava 4 4 4 4Intra- Portal vein 4 4 4 4venous SERA prophylaxis 0 4 4 4

Oral SEBA 1131 0 0 0 2(intra- Free 1131 0 0 0 2gastric)

that the complex was not actively absorbed and tte protein-bound radio-

activity was due primarily to combination of I131 following absorptionwith nonspecific proteins in the blood. Thus, it seemed doubtful thatmeaningful organ localization of SEB could be obtained following oralchallenge of the iodinated toxin.

The inital IV challenge studies were reported to the Comissionlast spring. The results of thpse studies will be combined with recentlyacquired data; however a different format will be employed.

Percentage of total radioactivity injected is plotted against timein hours for the various body parts measured (Figure 2). Various organsare presented individually in a form that clearly demonstrates the changesthat occurred.

1. Carcass and blood.

Blood values were calculated on the basis of blood volumebeing equivalent to 5. of the animals live weight. Carcass values includeradioactivity of blood trapped in the carcass. Similar patterns wereseen following IV and portal challenges: there was an early increase ofapproximately 40 of administered radioactivity in the carcass mainly dueto the high percentage in the blood. This was followed by a prompt fallfrom the early peak in both carcass and blood. At 30 min the carcassvalues began returning toward the amount of the early peak and slightlysurpassed this value at 6 hr. This was attributed to tissue absorptionand not solely to the amount in the blood. The blood follows a disappear-ance curve in complete agreement with the data repotted by Captain Rapoportwith the slight late increase due to circulating free I

With antitoxin prophylaxis an entirely different picture wasseen. Throughout the 6-hr period, carcass levels were between 75-80% ofadministered radioactivity. The blood level was slightly higher than the

37

I ZI,-" '..

FIGURE I. BLOOD ACTIVITY FOLLOWING ORAL ADMINISTRATIONOF SEB- 1131.

I

38

IV DOIThL AMTr0IN

00t LU*fA

4 IVa- a.

Tldft IN ova TIE 1 O R

j00t

4L,4 iA

Ito

a I V 4* * 0 4

gO' A bigH Ah ITOXI M4 TYRr, I 1

TIMEA1OI SY THHOUROSVNACVA(v

Ais STOMACH VEIN ANDESA A~TN THYROLAID

VIE ; " "

FIGURE 2. DISTRIBUTION OF SES-1131 FOLLOWING IV'

INOCULATION BY THE INFERIOR VENA COIA(Iv

AND PORTAL VEIN AND ANTITOXIN PR OPHWLAXIS.

39

carcass level until 4 hr. This was due to radioactivity of blood in variousorgans not counted with the carcass. The blood level showed a very gradualdecline in contrast to that observed in the monkeys challenged by IV orportal routes.

2. Lungs.

Again, values for the IV and portal routes were not significantlydifferent and the antitoxin data were in marked contrast. A peak value of8.9% was attained at 2 hr, more than twice as much as the IV value and 2times greater than the portal. At 6 hr the level was still more than twicethat of the other two. The reason for this difference was not readilyapparent. Retained blood with its high level of radioactivity after anti-toxin might have accounted for part of this difference. Perhaps binding oftoxin to white blood cells (WBC) and trapping of damaged WBC's in the lung(a concept presented by Captain Crawley at an earlier meeting) might havebeen accentuated in some manner by the presence of antitoxin. Anotier expla-nation may be that the toxin-antitoxin complex is taken up by pulmonary WBC'sor macrophages. It is possible that late pooling of blood in the lung, inassociation with pulmonary edema, could have accounted for our findings.

3. Liver.

After IV challenge there was a slightly higher percentage ofactivity than following portal challenge until 6 hr when the values of bothwere essentially the same. In contrast to the picture in the lungs, thepassively immunized animals showed a notable decrease in activity. At30 min both IV- and portal-challenged animals had a value of approximately20%, as compared to 10% for immunized animals. However, the IV and portalvalues fell steadily to a value of 6.5-6.7% at 6 hr; the antitoxin valuefell to approximately 7% at 2 hr, leveled off, and at 6 hr was similar tothe others. The absence of a larger accumulatio- f radioactivity in theliver after IV challenge with SBB-11 31 seemed to icrease its importanceas the major organ of SIB removal from blood.

4. Kidneys and urine.

Kidney values revealed a very interesting picture. With anti-toxin, very little activity was found at any time throughout the 6 hr.Although boch the poxtal and IV values were maximal at 30 min and steadilydeclined to < 10% at 6 hr, there were statistically significant differencesbetween them at both the 30- and 120-min levels.

The urine findings indicated the rate of deiodination of SEB-1 13 1

and excretion of the label in the urine. It should be noted that in neitherthe IV or the portal challenged groups did the urine and kidney values at6 hr come close to equalling the kidney values at the 30-min peak. Thisimplieid that the initially high concentrations of radioactivity had left

the kidney via the bloodstream rather than via the urine.

40

5. Stomach and intestinei.

The stomach and intestines were gronid to. ner sin.e theactivity found here was due primarily to free I that had been excretedinto the gastrointestinal tract. This gut localization of free iodide iswell known.!/ Note that prophylactic antitoxin which markedly lesseneddeio Mation of the toxin was associated with a much lower gut accumulationof I

6. Heart, thyroid, spleen and brain.

Radioactivity was very low in these organs. Values for heart,spleen and thyroid were similar. In the brain the greater rise in theantitoxin group was due to the large amount of activity present in theblood that bathes the brain. The values were low and probably of littlesignificance. Since the spleen is an important site of reticuloendothelialactivity, the low percentages found here have helped in eliminating thereticuloendothelial system as important in the hmndling of 3B. Theincrease seen in the thyroid represented uptake oZ free I

From this data and that of the previous presentations by Drs. Rapoportand Hodoval, it is evident that the antitoxin is very effective in preventingthe clearance of SIB from the bloodstream, i.e., greatly affecting itsdistribution among the various organs. The antitoxin-SEB-1 13 1 aggregateis shown to be fairly stable. By preventing the uptake of SIB by the organs,the toxic effect of SEB is combated completely or greatly delayed.

From these distribution studies, the kidneys primarily and the liversecondarily appear :o be the major organs engaged in actively localizing,deiodir.6ting, and p4ssibly detoxifying the SEB. The fact that antitoxinreduces the amount of SEB found in these organs, especially in the kidney,supports this contention. Another thought is that the kidney is used asa storage site for SEB to be released later and as a deiodination pointwith the free 1131 being returned into the bloodstream. The return offree 1131 is evident by the low quantities in the urine and the increasein the gastrointestinal tract.

The increase in lung radioactivity following antitoxin prophylaxis isdifficult to explain.

Further work is planned, aimed primarily at the kidney, to distinquish

the role of this organ in the handling of SEB.

SUMMARY

Oral, IV and portal routes of administration of SEB-1 131 were studied.The oral route is of no value because of destruction of the complex uponabsorption through the intestinal wall. The IV and portal routes are oZvalue but there is no real difference between them.

I-

41

Antitoxin is effective in preventing localization of toxin.

The major orgar- concerned with the handling of US, at least in theearly stages during its very rapid disappearance from the circulating blood,are the kidneys and secondly the liver.

LIT UATURI CITeD

1. Albert, A., and R. F. Keating, Jr. 1952. The role of the gastro-intest$Vnal tract, including the liver, in the metabolLza of radiothyroxine.Endocrinology 51:427-443.

P 6.4I

42

EFFECTS OF STAPHYLOCOCCAL ENTEROTOXIN B ON THE COAGULATION MECHANISMAND LELqoCYrIC REKSPONSE IN BEAL DOGS --- A PRELIMINARY REPORT

Charles F. Gilbert, Captain, mC*

Much of the reported work with staphylococcal enterotoxin has beenin areas of investigaLion cther than that of blood coagulation. The onlyreports een relating to clotting were those of Sugiyama and co-workersin 196-7 and Bernheimer and Schwartr.2/ earlier this year. Sugiyama andhis group were mainly concerned with fibrinogen and platelet changes afterenterotoxin administration in rabbits and monkeys. Bernheimer and Schwartzdemonstrated the capacity of enterotoxin to injure or destroy rabbit plate-lets in vitro.

The finding of hemorrhages in dogs receiving staphylococcal entero-toxin B (SEB) suggested that a defect might be present in the hemostaticmechanism and indicated the need for i.,veaiigation of this area. Thebleeding present in dogs was cspecially proinent in gut, liver, andlymphoid tissue, including the lymph nodes, spleen, thymus and Peyer'spatches. In fact, enterocolitis and hemorrhagic manifestations were themost striking findings at autopsy.

Ten adult purebred beagle dogs were used; 9 received an intravenousinjection of 100 g/kg of highly purified SEB, and one, 50 pg/kg. Thetests employed in this study included whole blood clotting time, clotretraction and lysis, prothrombin! time and partial thromboplastin time.Total and differential whit blood cells counts were performed.

Blood (8 ml per sample) was obtained using plastic syringes from theexternal jugular vein. One sample was taken from each dog before toxinadministration; after challenge, 18 specimens were collected over a 51-day period.

For the coagulation studies, 9 parts of blood were treated with 1 partof 3.2% sodium citrate. The plasma was separated by centrifugation andwas either tested immediately or frozen at -20 C for subsequent testing.All coagulation studies were performed in non-coated glas . Blood for thetotal and differential leukocyte counts were processed by :.atdard tech-niques.

The whole blood c otting time was performed by a L, dification of theLee-White technique;'/ the prothrombin time by the method of Quick;_/ andthe partial thromboplastin time by the method of Langdell, Wagner, andBrinkhous.5/

* Pathology Division.

43

The second tubs of the clotting time teat was observed for clot

retraction 30 min after the blood was drawn and at hourly intervals there-after. The degree of retraction wa graded from 3+ to 0. All observationswere made by one Individual. After retraction, the tubes were held at roomtemperature and observed for 72 hr for evidence of clot lysis.

The range and mean of the prothrombin time end partial thromboplastintime were established for the controls end the 18 postchallenge bleedingperiod*.

All dogs became 1Il within 3 hr after toxin administration, as evidencedby emesis and, sometimes, diarrhea. Two dogs died, one at 13 hr and one at4 days postchallenge.

The prechallenge clotting time for the 10 dogs ranged from 6-l0 minwith a mean of 8. Figure 1 illustrates the average clotting time for eachbleeding period and shows the prolongation immediately after challenge andto 22 min on the 28th day. The clotting time returned to the upper normalrange by the 51st day. The alterations in the clotting times suggest theremay be a deficiency of one or more of the plasma clotting factors.

All clots retracted normally (3+) before challenge and 1 hr afterchellenge)(Figure 2). Beginning at 3 hr, however, there was diminishedretraction and this continued until the 2nd postchallsnge day when onlya slight degree of retraction was present in 2 of the samples. Thereafter,clot retraction gradually improved and was normal on the 23rd day. Thesefindings suggest that there may be a deficiency of fibrinogen or of plateletsor possibly a defect in platelet function.

There was no evidence of fibrinolysia in any of the samples during a72-hr observation period.

The range of the prothrombin time in the prechallenge samples was8.3-10.5 sec, the mean, 9.5 (Figure 3). Prothrombin, time decreasedslightly 1 and 3 hr after challenge. At 6 hr, there was a prolongationto about 12.5 sec. Thereafter, the prothrombin tiff returned to the normalrange, except on the 30th day when the values were slightly below theestablished range. The abncrmality in the 6-hr samplings also suggesteda deficiency of one or more of the plasma factors; this is supported bythe data on whole blood clotting time.

Prolongation of the partial thromboplastin time occurred at 6 hr; it

persisted until about the 8th postchallenge day when the values returnedto the established normal range (Figure 4). This too, in conjunction withthe clotting time and prothrombin time, is additional evidence for adeficiency of one or more of the plasma factors.

deiin&-noteprta hobpatntmeocre t6h;i

LI. .. ... I

1141.

amma.

Iliu

a.saf 0e lh"

45

W.Y

P046. VRIATION IN 4OOtNOIWN tom

Das 4 is so so040 SO

IVI-~M4 WAMMIS- 0 PAN71"

I464

One of the more dramatic changes associated with the toxin was therapidly developing leukopenia. As seen in Figure 5, the average controlleukocyte count was 16,000/rn. One hour after toxin administration, thewhite count decreased to 1,800; this low level persisted in the 3-hrsampling; thereafter the count increased to about 20,000 on the 4th day.The leukocytosis, though slight, continued until the 12th or 14th day.These findings are similar to the response that may occur in an over-whelming bacterial infection in which there is an initial leukopeniafollowed by leukocytosis. The increased count on the last day remainsunexplained.

The changes in the differential counts were also most marked immediatelyafter challenge (Figure 6). Before toxin administration there was anaverage of 70% neutrophils and about 30% lymphocytes. By 1 hr afterchallenge these cells were present in almost equal numbers; subsequentlythe neutrophils gradually increased and the lymphocytes decreased. Duringthe 1st day and continuing through the 3rd day, there was a slight "shiftto the left" with the appearance of 7-8% early forms. After the 4th daythe cells were present in normal numbers.

Immediately after toxin administration, there was an average of 9normoblasts among 100 leukocytes counted. During the remainder of thelot day, the normoblasts decreased and were absent during the remainderof the experiment.

The appearance of early forms of each series, i.e., normoblasts andbands, the leukocytosis following the leukopenia, indicate that themarrow was not impaired by the toxin and was capable of responding.Furthermore, the marrow was histologically normal in the 2 animals whichdied.

As mentioned earlier, one of the animals died in 13 hr and another in4 days. Complete autopsies were performed shortly after death. Therewere losses of about 20% of the body weight. In addition to severe entero-colitis, there was bleeding into the lymphoid tissue and liver. There wasno evidence of intravascular coagulation or infarction. Both deaths wereattributed to dehydration and hemorrhage with severe pneumonia in theanimal dying on day 4.

SUMMARY

Ten dogs were given staphylococcal enterotoxin B intravenously andwere observed and bled over a 51-day period. The whole blood clotting time,clot retraction, clot lysis, prothrombin time, partial thromboplastin time,and total and differential white blood counts were performed on each sample.An abnormality was indicated by each test at some point during the testingperiod with the exception of clot lysin. The most frequent and more severeabnormalities occurred within the first 4 days after challenge. The resultsindicate that alterations of the clotting mechanism occur after administration

47

LIUKOCYTIS

CT otovm j

to.

OAS

LIUNO"ILI

40

to

k...*COdit LAS?.

T INIOTOXI

F14- 6 WFghhNTIAI. LgYNOCYTS (OUNT

48

of staphylococcal tnterotoxin B and suggest that the plasma factors areprimarily involved. Also indicated by these studies is the need forfibrinogen determinations and platelet counts. Changes in the leukocyLicresponse appear to be similar to those which can occur during an over-whelming bacterial infection.

LITERATURE CITED

1. Sugiyama, H., I. H. McKissic, Jr., and T. Rayama. 1964. Hyperfibrino-genemia and thrombocytopenia after staphylococcal enterotoxin. Proc.Soc, Exp. Biol. Med. 117:726-729.

2. Bernheimer, A. W., and L. L. Schwartz. 1965. Effects of staphylo-coccal and other bacterial toxin. on platelets in vitro. J. Path.Bact. 89:209-223.

3. Lee, R. I., and P. D. White. 13. A clinical study of the coagu-lation time of blood. Am. J ed. Sci. 145:495-503.

4. Quick, A. J. 1945. On the quantitative estimation of prothrombin.Am. J. Clin. Path. 15:560-566.

5. Langdell, R. D., R. H. Wagner, and K. M. Brinkhaus. 1953. Effect ofantihemophilic factor on one-stage clotting tests. A presumptivi testfor hemophilia and a simple one-stage antihemophilic factor assay pro-cedure. J. Lab. Clin. Med. 41:637-647.

49

SEROLOGICAL MZTHODS FOR DITECION OF STAPHYLOCOCCAL ENTEROTOXIN B ANTIBODY

Martha K. Ward, Captain, USPHS*

When the Serology Section of the Bacteriology Division began itsoperation, the task assigned highest priority was that of the establishmentof a routine serological service for determination of staphylococcal entero-toxin B (SIB) 4ntibody. Most of the work repbrtcd here was done by MajorJohnnie Runnels.

Several serological procedures which were used originally for thetitration of toxin have been adapted by other workers for use in tests forenterotoxin antibody. Among these were the hemagglutination (HA), complementfixation (CF), and several agar-gel diffusion (AD) techniques.

Since a variety of serum specimens of interest to investigators had beenexamined in other laboratories by a HA technique, this procedure was selectedfor initial studies. The protocol used was identical with that employed inother laboratories at Fort Debrick. Sensitization of red blood cells (RBC)for the test was accomplished by coupling toxin to sheep erythrocytes withBis-diazotized Benzidine (BDB). It should be emphasised that each step ofthis procedure, from the original preparation of the BDB compound throughthe final washing of the sensitized cells, was extremely difficult to controland apparently very sensitive to minor variations in temperature and timingand perhaps other unknown factors.

It soon became apparent that while the BDB HA test appeared to be ahighly sensitive method for detection of antibody, wide variations in titerswere observed when the same serum was examined repeatedly. For example,in repeated tests on a large number of serum samples the variations in endpoints observed were as much as 8 tubes in a 2-fold dilution scheme in anumber of cases. The mean variation in titers of one group of 20 serarepeatedly run together in the same protocols was more than 4 tubes or16-fold. Only 2 of these varied as little as 2 tubes from day to day. Inaddition, about 57. of sera which were negative in one test would be positiveon another day with titers occasionally as high as 1:640.

In view of these results, a rather intensive examination of the testitself was begun. Recognizing that our personnel were largely inexperienced,we thoroughly checked and carefully controlled all of the obvious factorswhich could affect results of any serological determinations, i.e., accuracyof dilutions, timing and temperature in preparation of all reagents, as wellas during incubation of the test, the reproducibility of reading end points,etc. Possible variations in sheep RBC as related to time of storage, bleed-ings from different animals and repeated bleedings from the same animal were

* Bacteriology Division.

** Presently at the Department of Veterinary Science, University ofWisconsin, Madison, Wisconsin.

50

also considered. The results of these rather extensive efforts to controland standardize all factors involved in the BDB RA test all led to thesame conclusion: the major variable contributing to lack of reproduci-bility of the test was the procedure involving the BDB coupling of antigento the RBC. If one can judge from other agglutination procedures, it isreasonable to assume that both the number of cells coupled properly withantigen and the concentration of antigen per cell could affect the endpoint of the test. Since, ae has been pointed out by other workers, BDBis a highly active compound capable of reacting with a variety of chemicalgroupings in a totally unpredictable fashion, a theoretical considerationof the most difficult part of the test procedure to control also pointedto the coupling procedure.

Originally it was not known that other laboratories using this testhad had similar experiences. We only knew that "certain difficulties'had been encountered. Upon further, more searching, questioning it wasdiscovered that our experience was not unique, wherever the day to dayreproducibility of the teat had been examined. For example, Dr. Silvermanof the Biological Laboratories (in whose laboratory most of the earlywork with the BDB test at Fort Detrick had been done) had found variationsin titer of > 10-fold when a single serum was examined with differentantigen preparations. Both Dr. Silverman and personnel of the BacteriologyDivision arrived at the conclusion that in its present state of developmentthe BDB HA test is of limited usefulness for quantitative studies ofantibody to SEB, despite its great sensitivity and its, at least assumed,specificity.

There was no alternative but to look elsewhere for a good serologicalmethod or combination of tests which could be used for detection ofantibody and which also could be quantitated accurately for studies ofthe immune response Lo antigen and imune status of particular individuals.Silverman's group has begun studies on a CF test and ts preparing toinitiate work with several other techniques. Dr. McCann of th-is Divisionhas been working for some time on an AD test (see the following paper);the Serology Section has done some preliminary work on a HA techniqueemploying antigen adsorbed on tannic acid-treated erythrocytes. It isour hope that with close cooperation and collaboration among these severallaboratories the basic problem will soon be resolved.

Limited data comparing the results of 2 or more of these tests onthe same serum specimens are available and have been summarized in severaltables. Table I shows titers obtained in repeated tests on hyperinuunehorse serum with the 2 HA techniques. RBC from different bleedings of the Isame sheep were used for antigen preparations after varying periods ofstorage. It is readily apparent that the tanned cell technique is markedlyless sensitive than the BDB test as Judged by end point titer. However, inthese trials there is no significant variation from preparation to prepa-ration with the tanned cell test while the variation in titer with the BDBtest is from 1.300,000 to 1:20,000,000, or 6 tubes in a 2-fold dilutionscheme, on days when the 2 tests were run in parallel.

A

51

TADLE I. VARIATION IN UMAOGLUTIMATIO TITERS USING RBC FROMDIFFERENT BIDING8 OF SM SEEP

RECIPROCAL HA TITERY'DATE BLOOD DRAWN& DATE BLOOD USED DRS. TSI/

7 July 5,000,000 10,00027 June 9 July 5,000,000

12 July 600,000 10,000

12 July 20,000,000 10,0007 July 13 July 600,000 20,000

14 July 300,000 10,000

14 July 20 July 160,000

26 July 300,00021 July 27 July 2,600,000 20,000

30 July 10,000

a. 100 ml of blood drawn from sheep No. 4270 on dates shownabove.

b. Hyperimmune horse serum. Reciprocal of last dilutiongiving agglutination.

c. BDB equals hemagglutination with Bis-diazotized Denezidinecoupled antigen.

d. TS equals hemagglutination with antigen absorbed on tannedsheep RBC.

In Table II results of the 2 HA tests and the AD test on serum samplesselected at random from 200 individuals are summarized.

The upper portion of the table shows the number and percent of samplespositive by each of these techniques. The lower portion compares theresults of the AD test with the 2 H& techniques. There is little corre-lation between detectable precipitating antibody and BDB HA titers under1:512, or tanned cell HA titers under 1:20. Much better correlation is

t .. ........ = I

52

TABLE Zl. SOMIARY OF R&SULTS OF 3 SCROLOGICAL TESTS ON SAMPLRSFROM INDIVIDU SLECTED ATRANDOM

TEST NO. POSITIVE % POSITIVE

BDI 153 77TSb / 65 33Agar diffusion 55 28

AGAR DIFFUSION

TITK%/ NO. SAMPLESNo. Positive % Positive

BDB

Negative 47 4 8.5< 512 102 15 15.0

512 23 14 61.0> 512 26 22 79.0

TS

Negative 124 16 13.010 - 20 37 15 40.540 - 160 23 15 65.0320 - 1280 6 6 100.0

a. BDB E.uals hemagglutination with Bis-diatotized Benzidinecoupled antigen. 200 samples tested.

b. TS equals hemagglutination with antigen absorbed on tannedsheep RBC. 190 samples tested.

c. Reciprocal.

seen at higner titers in the HA tests. In Table II results of the CFtests, performed by Dr. Silverman's laboratory, as well as those obtainedby the 3 techniques used in the Medical Unit laboratories, are tabulated,for serum samples takenfrom 15 individuals with known exposure to SEB.The samples examined were taken 16 weeks after exposure in all cases. Twopoints of interest are apparent from the table: (1) there is a close corre-lation between the results of the AD and CF tests and (2) there is a ratherconsistent relationship between the results of the tanned cell HA test andthose of the AP test, the titers of the former being 10- to 20-fold higher

53

TABLE III. RESULTS OF 4 SEROLOGICAL TESTSS / ON SAMPLES rROM15 INDIVIDUALS WITH KNOWN EXPOSURE TO ENTCROTOXIN

RECIPROCAL TITER

SUBJECTBDB CF AD TS

1 800,000 64 32 12802 200,000 32 32 3203 400,000 32 16 1604 40,000 16 8 3205 16,000 16 8 1606 20,000 16 4 807 40,000 8 4 408 10,000 8 4 409 1,000 Neg Neg 2010 8,000 16 8 8011 16,000 16 8 8012 16,000 8 8 8013 4,000 Neg Neg 2014 8,000 AC 2 4015 2,000 Neg 1 20

a. BDB equals hemagglutination with Bis-diazotized Benzidinecoupled antigen.

CF equals complement fixation (tests performed by Dr. Silverman'slaboratory).

AD equals agar diffusion.

TS equals hemagglutination with antigen absorbed on tanned sheepRBC.

in every case. On the other hand, there is a much less constant relationshipbetween the titers of the BDB HA tests and those obtained by the other methods.

The final evaluation of any serological method muat cake into considerationthe question of whether or not the results of the test provide any useful infor-mation related to previous experience with the antigen, and the immune status(or immune response to antigen) of the individual whose serum is being examined.Definitive information on this point for any of these methods must awaitfurther work carefully planned for this particular purpose.

At present only suggestive information is available. If we assume thatthe 200 individuals, referred to earlier, constitute a typical sampling, itappears that from 28 to 77% of the population have had some experience withenterotoxin antigen, the number depending upon the test used.

•I

54

The question of whether any of the available in vitro methods toevaluate potential protection against challenge provi--T-by detectableantibody can be used Is impossible to answer at present. There has beenno opportunity to go through all of the records available on the largenumber of monkeys which have been used in a variety of studies withenterotoxin. However, data have been examined carefully on 184 monkeyschallenged by the intravenous route with known doses of SEB varying from0.2-300 pg/kg of body weight. Unfortunately, for purposes of analysis,prechallenge antibody levels were determined after the fact, and thechance distribution of animals with antibody was such that most of themwere in the lower dose groups. Of 40 animals challenged with 300 .g/kgnone had antibody detectable by the AD technique and only 4 had prechal-lenge antibody detectable by the BDB HA test. The remaining 30 animalswith HA antibody and the 16 having precipitating antibody were distributedmore or less equally among the 100, 30, 10 and 0.2 pg/kg-dose groups.

Upon further examination of the data, differences between death andsurvival for animals with and without antibody were calculated for animalschallenged with 100 and 30 pg/kg as a single group since in these 2 casesmore than 50% of all animals succumbed to challenge. Similarly, differ-ences between illness and no illness for animals with and without antibodywere calculated for all animals in the 10 and 0.2 pg/kg-dose groups, sincefew animals died, but more than 50% became ill after challenge with thesedoses.

Although the number of animals with antibody was too small to draw anyfar-reaching conclusions, the results of this analysis showed that therewas a significant difference between the responses of animals with andwithout antibody in both cases. The level of significance was much higherin the calculations made for animals having precipitating antibody than forthose having HA antibody. It is noteworthy that none of 8 animals withprecipitating antibody succumbed to the 100 or 30 ig doses while 5 of 15animals with HA antibody and 41 of 65 without antibody detectable by eithermethod died as a result of challenge. Similarly, only 1 of 8 animals havingprecipitating antibody became ill after challenge with 10 or 0.2 Pg/kg. Theillness rate amcng those having HA antibody was 6 of 15 and tha amongmonkeys without antibody detectable by either method was 36 of 46.

Despite the small numbers involved, these data suggest that thepresence of precipitating or HA antibodies affords some protection againstthe lethal effect of SEB at relatively high doses, and may also affordsome protection against illness, at doses of toxin well below the medianlethal dose but in the range of the illness dose50.

55

StI@4RY

The problem of aerological tests for the detection of SXB antibodies isstill unresolved. While the widely used BDB HA test appears to be more highlysensitive than any others employed thus far, it is so difficult to controlthat attempts to quantitate it for routine work appear doomed to failure.The less sensitive tanned cell HA, CF, and AD methods appear to be infinitelymore reproducible, reliable and amenable to quantitation. It is impossibleto say at the moment which, or whether, any of these technique& will proveuseful in estimating protection against challenge. It appears that antibodydetectable by agar gel diffusion is more closely related to protective anti-body than that demonstrated by the BDB HA test.

I

t

56

DETECTION AND MEASUREMNT OF IMMUNOLOGICAL

kESPONSES TO STAPHYLOCOCCAL ENTEROTOXIN B

Virginia G. McGann, Ph.D.*

During the initial stages of the staphylococca! enterotoxin B (SEB)program, serclogic methods were developed by other investigators for monitoring production and purification of toxin. The methods, however, werenot entirely satisfactory for quantitation or analysis of antibody. Conse-quently, studies were initiated to investigate in vitro methods for detection,measurement and analysis of the antibody response to vaccination and challenge.This report summarizes our approach to these problems.

Exploratory experiments with hyperinmmune horse and rabbit se a suggestedthe feasibility of using Oucht'rlony agar diffusion techniques.! / Withhyperimmune sera, reproducible titers were obtained by testing increasingdilutions of serum against a constant amount of antigen, and the titersseemed to reflect antibody potency. Several problems appeared when pro-cedures developed for hyperimmune sera were used to examine sera fromanimals with less enterotoxin experience. The major difficulty was theinhibitory effect of excess antigen or antibody on immune precipitation.Excess antigen completely inhibited formation of immune precipitates withsera from all species, and excess antibody in sera of man, monkey, chimpanzeeand horse markedly inhibited precipitation. The optimal amount of toxin fortitrations of strong sera was 5 g, wheceas for many human sera it was 0.2 [ig.Consequently, it was necessary to test unknown sera against several concen-trations of toxin, a procedure that made difficult a meaningful estLmate ofantibody potency.

To avoid this difficulty, a method was developed for demonstrating thezone of antigen-antibody equivalence by Ouchterlony techniques (Figure 1).The procedure consisted of mixing increasing concentrations of toxin withconstant amounts of serum in a series of tubes. After incubation the mix-tures were added in sequence to the center row of wells. Reference toxinwas added to each well in the top row and reference antiserum to each wellin the bottom row. After 1-3 days, plates were examined for location ofimmune precipitation. If a mixture contained excess antibody, n precipitateformed I h the reference toxin; if axcess toxin, precipitate tormed withreference .ntiserum. The neutralization zone was the area in which neithercomponent was in excess. End-points of the zone were affected by incubationtime and concentration of reference material. The mid-point, called theequivalence point, was quite insensitive to these influences; by conversionto ag toxin/ml serum it could be used to compare potency of a wide range ofsara. For a hyperimmune horse serum, this method gave essentially the sameequivalence value as did quantitative precipitation. When this serum was

* Bacteriology Division.

57

FIGURE 1. AGAR DIFFUSION TEC14NIQUE FOR EQUIVALENCEDETERMINATIONS. (Top ROW- REFERENCE TOXIN; MIDDLEROW, SERUM-TOXM MIXTURES; SC1'TOM ROW, REFTIENCEANTISERUM, )

58

absorbed with toxin at equivalence, no free toxin or antibody could bedemonstrated in the supernatant fluid, and tests oi it for prophylacticefficacy in monkeys indicated removal of protectiv, antibody.

Serum sawplea £rw severs! speLies of animals that had received avariety of experimental treatment were tested. Equivalance values rangedfrom 1-2800 pg toxin/ml serum. The relationship between equivalence pointand agar diffusion titer, as measured with I pg toxin, was examined todetermine whether titers could be used as a measure of antibody contentwhen there was insufficient serum for equivalence determination (Figure 2).If the loglo equivalence point waa plotted against the loglo reciprocaltiter, a fairly good straight-line relationship was obtained. The reasonfor the difference in slope for horse saera, as compared with monkey andhuman sera, is not known. It may reflect the type of exposure to entero-toxin or the influence of species-specific, non-antibody serum factors.Another possibility that will be discussed later is the nature of theantigen-antibody reaction.

Limited data suggest a relationship between equivalence point andprotective activity. Studies with mon'eys challenged by the intravenousroute indicated that with high challeuge doses (300 pg/kg) the equivalencepoint may measure protection against lethal effects of enterotoxin, butnot against illness. Several sera and serum concentrates used to treatmonkeys after challenge with enterotoxin were available for study. Table Isummarizes data on the fate of challenged monkeys after treatment withantiserum at, or near, the equivalence point. "Survival" indicates thatall animals in ,be test group survived challenge; "death," that at leastone succumbed. For example, with the 0.00067 ml/pg dose of Lot 19575,1 of 6 died, whereas 6 of 6 controls died. There was good correspondencebrtween equivalence value and the dose of antitoxin that prevented lethality.

In other studies 3 groups of 6 monkeys each were immunized with a toxin-antitoxin mixture at equivalence, or with the precipitate or supernatantfluid from the mixture. At time of challenge all animals had circulatingantibody with equivalence values that ranged from < 1-60. After challengewith 300 pg/kg, 17 of 18 animals became ill, but all survived. Mediantime to illness for control monkeys and for animals with equivalence values< 10 was approximately 1 hr; for those with values of > 10, almost 3 hr.The animal that did not become ill was in the latter group. When sera frommonkeys that had no known prior experience with enterotoxin were examined,approximately 9% had detectable levels of precipitating antibody, rangingfrom amounts too weak to quantitate to titers of 1:8, all representing anequivalence of < 10. After challenge with varying doses of toxin, allanimals with prechallenge antibody survived, whereas only 50% of thosewithout detectable antibody survived.

Serial bleedings were obtained from 25 individuals known to have beenexposed to type B enterotoxin. Twelve had participated in an ingestionstudy, and approximately I year later 2 of the 12 were among a group of

59

xxx 0

0 ci 0

21 Eo 0.Z ~ 0. 1

4 C)0 Z -

a CLZ

(nZ.0 00

N LaR

IrI

60

TABLE 1. COMPARISON OF NEUTRALIZATION ACTIVITY OF SERA WITH EFFICACYFOR PASSIVE PROTECTION OF MONKEYS

SERUM SAMPLd EQUIVALENCE POINT SERUM DOSE MONKEY(ml serum/pg toxin) (ml/ g toxin) RESPONSE

SEBAI / 0.002 0.003 Survival0.001 Death

SAEB Lot 19575 ! 0.0004 0.002 Survival0.00067 Death

Human -g1obulin 0.01 0.02 Survival

Human serum 0.1 0.1 Death

Monkey pool 0.04 0.01 Death

a. Staphylococcal enterotoxin B antitoxin, Lot 1 (equine origin,

U. S. Army Medical Unit preparation).

b. Staphylococcal anti-enterotoxin B, The National Drug Co. preparation.

i5 individuals involved in an accidental aerosol exposure. In the ingestion

experiment, subjects were fed purified toxin. Seven individuals had nodetectable antibody before exposure; 6 of 7 responded to toxin, 5 with overtillness and 1 with a low level of antibody. Five individuals with pre-

exposure titers showed no toxic symptoms; 2 of the 5 had high preexposuretiters that remainet' unchanged; the other individuals had barely detectablepreexposure antibody that increased significantly within 2 weeks afterexposure. Two of the group with preexposure titers were subsequentlyinvolved in the accidental aerosol exposure to an unknown concentration

of enterotoxn. They werp ALH, whose antibody had disappeareu trom circu-lation, and EWG, whose antibody level remained unchanged. Figure 3 showsthe antibody responses of the 15 subjects involved in the incident. No

antibody could be detected in sera from admission bleedings of the hospital-ized individuals, but REN, who was ill and not hospitalized, had a low anti-body titer. Except for JFF, all who became ill. had an antibody response withthe maximum at 4-10 weeks and no significant change between 4 and 9 months

after exposure. Of 5 individuals who did not become ill, 2 had significantlevels of antibody that remained unchanged; 2 showed a weak response, andone never developed antibody.

61

tA(D W

.00

L

60

CY "

03ZI'Ivuin3NI IWNXO f

62

A retent survey of 200 people selected at random indicated thst 55individuals, or 28% of the population, had precipitating antibody to SEB;

29 samples had sufficient antibody fer quantitation. An unexpected 9baeC-

vation was the sporadic appearance of antibody in sera from serial bleedings

of both man and monkey. When 4 bleedings of 10 individuals with no known

experience with toxin were performed at weekly intervals, antibody was

detected in the 2nd and 4th bleedings of one individual, and the 1st, 3rd

and 4th bleedings of another. In bleedings from one monkey, antibody was

present in one serum sample, but not in a sample taken 2 days later. Thispattern of response occurred only with sera that were too weak for quantAtation and may represent exposure to minimal antigenic stimulation. For

example, with oral challenge, one individual, who showed no overt response,

had weak antibody 2 weeks postchallenge but none at 4 weeks.

Throughout these studies there has been the suggestion that more than

one antigen.-antibody system was involved. With hyperimmune sera clear-cutevidence was difficult to obtain, but with human and monkey sera 2 antigen-

antibody systems frequently were suspected from examination of detection

tests and titrations. in the simple detection test 0.1 ml of unknown serumwas tested againet several concentrations of toxin. Sera too weak forquantitation formed only one line of precipitate, but many stronger seraformed 2 distinct lines, as shown in Figure 4. When saera taken before and

after challenge were compared 'Figure 5), a 2nd antibody could be demon-

strated in the postexposure sample. With the postexposure serum a double-

line reaction occurred, and one of the lines crossed the line common tothe pre- and postchallenge samples. In neutralization titrqtions 2 anti-bodies might be present, one requiring more toxin for ne- :alization than

the other.

Studies using disc electrophoresis are helping to interpret some of

these findings. Purified toxin was run in a large-pore gel in a cationic

system. The column was sectioned lengthwise; one-half was stained forprotein and the other was tested by immunodiffusion methods. Figure 6 is

a diagrammatic representation, comparing the stained section with theimmunodiffusion reaction against hyperimmune horse antiserum and againstpostexposure serum from an individual involved in the aerosol incident.

Shaded areas indicate diffuse reaction. A major and a minor protein

appeared in the stained section. At least 2 other minor fractions weredemonstrated by immunodiffusion. The rapidly-moving material that reactedwith horse serum appeared with a sample of 10 pg. Both components that

corresponded to the stained areas reacted with horse serum at > 5 4g of

toxin, but only one reacted with 1 4g. The slow-moving material was not

demonstrated with horse serum but was the only component that reacted withhuman serum at the 10-jg level, and the reaction occurred close to the

serum slit. With 5 jig, the reaction with human serum appeared as is shownhere. the slow-moving component forming a heavy, distinc* line and the

stained-area component forming a weak, diffuse precipitate. With I jig,only the stained-area material showed a reaction forming a heavy line Cfprectpitate about half-way between the column and the antibody slit.

63

'- T(

FIGURE 4. DETE=CTiON TEST FOR ANT:BODY, (T.TO~lN; S,$ERUM.)

64

FIGURE 5. ANTIBODY IN PRE- AND POST-EXPOSURE SERUMS.(7T,iN: .CQW0TR0L SERUM;S5-1- PRE-EXPO.INE SERUM;S-2- POST-EXPOSJRE SERUM.)

65

0 wZ ~0

o2z 0!

ZOZ

Zo z

wwp U~C) (D0

W W I-.

0

IM

66

This component identified with 1 of the 2 horse serum reactions but becauseof concentration effects we are not yet certain which one. In many humanserum samples the major antibody appeared to be directed against the slow-moving component. It is hoped that studies now in progress may clarify thesignificance of these results with-regard to protection. The results, how-ever, are compatible with work reported several years ago by other investi-gators. At that time 2 major and 1-2 minor components were demonstrated inpurified toxin by starch electrophoresis and fluorescein paper. Ouchterlonytests with hyperimmune rabbit sera indicated that although 3 componentscould be detected, one constituted over 99% of the toxin.

Paralleling the preceding experiments, investigations have been inprogress on the nature and purification of antibody. Immune electrophoreticstudies demonstrated that the antibody in human, monkey and rabbit sera wasa -'2 -globulin; in contrast in hyperimmune horse sera the major antibody waspresent in the T-globulin region, and only small amounts were in the 2-fraction. During the course of immunizing the horses, there was littledifference in titer of sera obtained 7 days after cumulative doses of either78 mg or 9 gm toxin, but during this interval the major antibody-containingfraction changed from the 2-- to the T-globulin region, and with 2 of 4animals y 2-antibody disappeared completely. Concentrates of this pooledh perimmune plasma, prepared by The National Drug Co. from a 28-50%(NH4 )2SO4 cut, showed apprcximately 5-fold purification, in terms of equi-valence values of g serum protein/4g toxin, but they contained 8-9 serumfactors in addition to antibody. After pepsin digestion only 5 nonantibodyfractions could be detected.

Investigations in progress have been directed toward increasing thepurity of plasma concentrates. Pilot studies suggested that salting-infractionation resulted in a product with essentia.Lly the same neutralizingactivity as The National Drug Co. product, but with fewer nonantibodycomponents. Further purification could be effected by passage throughDEAE-Sephadex or hydroxyapetite.

SUMMARY

Agar diffusion techniques have been of considerable value for studieson detection, measurement and analysis of the antibody response to immuni-zation and challenge with staphylococcal enterotoxin B. Methods weredeveloped for measuring antigen-antibody equivalence and for estimatingequivalence values from simple titrations. Equivalence values appeared tocorrelate with protection against lethal effects of enterotoxin in monkeysand with protection against illness in man. Immunoelectrophoretic analysisindicated the presence of at least 2 minor antigens in the purified entero-toxin preparation. The major antibody reaction of some human and monkeysera appeared to be directed against one of the minor antigens. Consequently,it will be necessary to isolate the antigenic compopents in order to determinethe contribution of individual antibodies to serum titer and to protection

67

LITERATURE CITED

1.Ouchterlony, 0). 1958. Diffusion gel methods for immunologicalanalysis. Progr. Allerg. 5;1-78.

69-UNC ASSIFIER

Security ClasificationDOCUMEN"T CONTROL DATA - R&D

fsoa ,,~e*nr cgr, of title. Iody of 06baoe wwd i,,d.OM onmtstioni mus.t 6* snie.d wiAM th# o,..It mmlo~ I,. - I-dJI otiflINATII G ACTIvi'Y (Co."NO. .itho) ae PCPORT GaCu A I TY C LASSI.ICC ATION

U. S. Army Medical Unit UNCLASS IF IEDFort Detrick, Frederick, Maryland 21701 26 #"OuR

____ ____NA

3 REPORT TITLS9

SPECIAL REPORT TO THE COMMISSION ON EPIDEMIOLOGICAL SURVEY OF THE ARMED FORCESEPIDEMIOLOGICAL BOARD

4 OI5CAPTIVE NO'ES (Typo .1 NPor And Inohtul" data*)

Special Report, Annual Meeting, Commission on Epidemiological Survey, 10 Sep 655 AU1S405tS) (Lost name,. laist name.. intial.)

CAROZZA, JR., F. A., RAPOPORT, M. I., HODOVAL, L. F., MORRIS, E. L., GILBERT, C. FWARD, M. K., McGKNN. V. G.

4 REPORT OATE J7, TOTAL NO Oil 010*18 1b NO 01 or.,

March 1966 70 9St CONTRACT ONt 6RANT NO 5 11~N 9P 009 tttOS9(S)

PROjec- l B533O1D164 03

CS lib o- ata oo($) (Any otat nIoabef tat may be asoItpod

tO AVA ILABILITY/LIUIATION NOTICES All distribution of this report is controlled. Eachtransmittal of this dociument outside the Department of Defense must have priorapproval of the Commanding General, U. S. Army Medical Research and DevelopmentCommand, or the Comanding Officer, U. S. Army Medical Unit, Fort Detrick,

I I SUPOPL RMEITARY NOTTS I.SPONSORING MILITARY ACTIVITYS. Army Medical Research and Development

omand, Office of The Surgeon General,partment of the Army, Washingto 26AS C.

11 ABSTRACT

A reviewa of the research program directed towiard determining the biologicaleffects of staphylococcal enterotoxin B is presented.

FORI A 7' UNCI.ASS IFIEDSecurity Classification

70-UNCLASSIFIED

4 LINK A LINK 9 LINK C

Interotoxin 'i OO .i.J

Biological effectsI

Immtuno logySerologyPyrogensTherapyI

INST)&tC TlONS

1. ORIGINATING ACTIVITY- Enter the name and address 1, ,AVAILABI1LITY,;LIMITATlON4 NOTICES: Enter any 11i1-of the contr a ctor. subcontractor, grantee, Department of De- Is ttons on lurther dissemination of the report, other than those

thenrepo.i~~ fhrognzto cror uhr sun imposed by security tassaificalol, using standard statementsthe report.such*

2: REPORT SECURTY CLASSIFICATION: Enter the ot~' l'ullidrqetrsmyoti cpe fti1l. security classification of the report. Indicate whether (1 "Qaiedrqsts yobincpsorhs

-'Re,.stricted Data" is included. Marking is to b e in accord' report from DDC."

an ce wmjth appropriate security rcgtulations. (2) "Foreign anntouncement and dissemination of this

2h GROUP: Automatic downgradlrtg is specif, le in 1.Di report by DDC is rot authorized."

rective 5200. 10 and Armed Forces Industrial Manual. Entteir (3) "U. S. Covernment agencies m'ay obtain copies oft he group n umber, Also. when applicable. show that optional I this report directly from DOC. Other qualified DDCnia~kings have been used for Group 3 and Group 4 as Author- I users shall retquest throuighised. .1

3. REPORT TITLE: Enter the cornillete report title in all (4) "U.l S. military agencies stay obtain copies of thiscapital falttrs Titles in all cases should be unclassified, report directly from DDC. Other qualified usersIf a meaningful title cannot be selected without clSaifica- shall request throughtton. show title classification In all capitals in parenthesisimmediately following the* title. __________________

4. DESCRIPTIVE NOTE& If appropriate, enter the type of (5) "All distribution of this report is controlled. Qlual-report. e.g., interim. progress, summary, annual, or final. tried DDC users shall request throughGire the inclusive dales when a specific reporting Period iscovered.

If the report has been furnished to the Office of TechnicalS. AUTHOR(S). Enie-r the namea(s) of authorWs as shown an Services, Department of Commerce. for sale to the public. indi-or in the report. Enter last name. fisat name, middle Initial. cate this fact and enter the price, it known.If military, show rardi and branch of service. The name of 1,SPLMNAYNTS s o diinlepasthe princtpal author is an absolute minimum requirement. II.y nsPL. NAYNTS s o diinlepaa6. REPORT DATE. Enter the date of the report as day,. oyntsmonth. year; or month, year. It more than one date appears 12. SPONSORING MILITARY ACTIVITY: Enter the name ofon the report, use date of publicat ion, the departmental project office or laboratory sponsoring (Pat-

7a. 110TL NUBER F PAES:ing !or) the research and development. Include address.?a OA UBRO AE:The total page count 13AITATEtra btatgvnabieadfcul

should follow normal pagination procedures, i.e., enter the 13ASRATEtraabrctiinabieadfctlnumber of pages containing informations. ..rvMary of the document indiacative of the report. ecen the lgh

.t mo a ,a !0 SPPrar elsevhere iv. the body of the technical te76. UM1ER F REFRECESEnte th toal nmbe of port. If additional space is required, a continuation sheet

references cited in the reiort. shall be attached.

LA& CONTRACT OR GRANT NUMBER: If atiproprilate, enter It in highly desirable that the abstract of classified re-the applicable number of the contract or grant under which ports tie unclassified. Each paragrasph of the abetract SthatIthe report was written, Iend with an indication of the military security classification

85, S&, Il 8id. PROJECT NUMBER: Enter the appropiriate of the inforniation in the paragraph, represented -a (TS), (5).

military departmnentl identification, such as project number. (C). or (UI)subitroject number, system numbers, task number, etc. There is no limitation no the lcngth of the abstract. How-

9a. ORIGINATOR'S REPORT NUMBER(S): Enter the offi. ir-er. the suggesited lenigth is from ISO to 225 words.cial report number by which the document will be identified 14. KEY W.ORDS: Key words are technically meaningful termtsand controlled by the originating activity. This number must or atiort phrases that characterize a report and may be used asbe unique to this report. index entries for cataloging: the report. Key words most be

9b. OTHER REPORT NUMBER(S): If the report has been selecterl so that no security classification is required. Iden-

assigned any other report numbere (either by the origirnator tielrs, such as equipm 'nt model designation, trade namne. 'II.or by rte spoonsor). also enter tie number(*). t, project code same. geographic location. may be used as

it,, word, but will be followei by an ind-icao of technical-11 ,net The assignmnt of links, rules, And weights in

UNCLASS IFIEDSeurly aSsificution